期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2007
卷号：XXXVI-4/W54
页码：91-93
出版社：Copernicus Publications
摘要：The increasing number of Earth observation sensors on the market is making the choice of optimized data sources to serve operational needs more complex than ever before. Applying a geographic perspective before selecting data sources and procuring services that offer comprehensive data source kits are two effective responses to this new trend. The paper presents a methodology newly developed and implemented in the field of building multi-scale vector databases on a national or regional size. One of the main problems when a vector database is to be built is to identify, select, and qualify the adequate input source data, knowing the classes of vector / objects to be captured, and the correspondent technical requirements (eg. minimum size, geometric accuracy, maximum "age" of data,…). The paper shows how a rigorous methodology based upon "traditional" geographical considerations, associated with a focused exploration of the catalogues of image providers, greatly help to select an efficient and as-cheap-as-possible multi-resolution portfolio of input data aimed at minimizing the risks of data capture as well as the QC tasks. The associated presentation will detail the methodology for the selection of pyramidal multi resolution image layers, and their qualification for vector production, vs the specific scales required. Finally, the authors would like to share the lessons drawn after several years about the technical specifications / requirements listed of the vectors themselves, for new bases as well as for updating existing ones, and bring these few pragmatic elements to the scientific community in charge of developing innovative algorithms for the new generation of tools for vector identification and capture
关键词：Multi-source; multi-resolution; vector database; data Kits; imagery; requirements and constraints
